Professional Documents
Culture Documents
Keywords: experience marketing, brand experience, customer experience management, scale development,
marketing communications
ow do consumers experience a brand? How is brand brand attachment) and develop a scale that can measure the
Brand Experience / 53
(Park and MacInnis 2006; Thomson, MacInnis, and Park Cognitive Science
2005). In contrast to brand attachment, brand experience is Cognitive scientists have investigated mental modules, or
not an emotional relationship concept. As we described pre- special purpose computational systems that respond to spe-
viously, experiences are sensations, feelings, cognitions, cific environmental cues and solve a restricted class of
and behavioral responses evoked by brand-related stimuli. problems (Fodor 1998). Pinker (1997) identifies four men-
Over time, brand experiences may result in emotional tal modules that correspond closely to the experiences that
bonds, but emotions are only one internal outcome of the Dewey (1922, 1925) postulates: sensory perception, feel-
stimulation that evokes experiences. ings and emotions, creativity and reasoning, and social rela-
As with brand attachment, customer delight is charac- tionships. However, Pinker does not list a separate doing
terized by arousal and positive affect; it can be considered module; he considers bodily experiences and motor actions
the affective component of satisfaction (Oliver, Rust, and and behaviors part of the sensory-motor module, similar
Varki 1997). Customer delight results from disconfirming, to Lakoff and Johnsons (1999) concept of embodied
surprising consumption (Oliver, Rust, and Varki 1997). In cognition.
contrast to customer delight, brand experiences do not
Applied Writings on Experience Marketing and
occur only after consumption; they occur whenever there is
Management
a direct or indirect interaction with the brand. Moreover, a
brand experience does not need to be surprising; it can be Pine and Gilmore (1999) study staged experiences in
both expected or unexpected. retail environments and events. For these settings, they
Finally, brand experience is distinct from brand associa- distinguish aesthetic (including visual, aural, olfactory, and
tions and brand image (Keller 1993). One of the most stud- tactile aspects), educational, entertaining, and escapist
experiences. Although Pine and Gilmores framework is
ied constructs of brand associations is brand personality
limited to retail settings and events, their four experience
(Aaker 1997). Consumers tend to endow brands with
dimensions nonetheless overlap with some of the dimen-
human characteristics that result in a brand personality,
sions of experience we discussed previously, thus confirm-
which consists of five dimensionssincerity, excitement, ing the need for conceptualizing a sensory/aesthetic, intel-
competence, sophistication, and ruggedness (Aaker 1997). lectual/educational, and emotional/entertaining dimension.
Brand personality is based on inferential processes (Johar, Schmitt (1999) proposes five experiences: sense, feel,
Sengupta, and Aaker 2005). That is, consumers are not sin- think, act, and relate. The sense experience includes aesthet-
cere or excited about the brand; they merely project these ics and sensory qualities. Consistent with recent research in
traits onto brands. In contrast, brand experiences are actual consumer behavior (Richins 1997), the feel experience
sensations, feelings, cognitions, and behavioral responses. includes moods and emotions. The think experience
Thus, because brand experience differs from brand evalua- includes convergent/analytical and divergent/imaginative
tions, involvement, attachment, and customer delight, brand thinking. The act experience refers to motor actions and
experience is also conceptually and empirically distinct behavioral experiences. Finally, the relate experience refers
from brand personality. to social experiences, such as relating to a reference group.
The five experiences are closely related to Deweys (1922,
1925) categorization, Dub and LeBels (2003) pleasure
Dimensions of Brand Experience construct, and Pinkers (1997) mental modules, especially if
To identify dimensions of brand experience, we next review we view Pinkers sensory module as including sensory and
pertinent work in philosophy, cognitive science, and applied behavioral components.
writings on experiential marketing and management. In this
literature, across a variety of disciplines, a fairly consistent Summary
set of experience dimensions, which are highly relevant to There has been considerable agreement in the categoriza-
brands, has been proposed. tion of experiences by philosophers, cognitive scientists,
and management thinkers. Therefore, for the scale develop-
Philosophical Investigations ment that follows, it is necessary to generate items along the
five experience dimensions that broadly emerged from our
The philosopher John Dewey (1922, 1925) views experi-
literature review: sensory, affective, intellectual, behavioral,
ence as the intertwining of human beings and their environ-
and social.
ments. He critiques the purely cognitive Kantian view of In line with our conceptualization, the experience
experience as knowledge, arguing that knowledge (classify- dimensions are evoked by brand-related stimuli (e.g., col-
ing, analyzing, and reasoning about things) is only one part ors, shapes, typefaces, designs, slogans, mascots, brand
of a persons understanding of the world. In addition to characters). Note that there is no one-to-one correspon-
intellectual experiences resulting from knowledge, experi- dence, such that a certain stimulus type would trigger a cer-
ences also include perceiving (through the senses), feeling, tain experience dimension and only that dimension. For
and doing. Moreover, human beings are fundamentally con- example, although colors, shapes, typefaces, and designs
nected with other people. Following Dewey, Dub and usually result in sensory experience, they may also result in
LeBel (2003) distinguish four pleasure dimensions: intel- emotions (e.g., red for Coca-Cola) or intellectual experi-
lectual, emotional, social, and physical pleasures. ences (e.g., when designs use complex patterns). Similarly,
Brand Experience / 55
TABLE 1
Descriptions of Experiential Brands
Abercrombie & Fitch Home Depot
Its a complete experience when you enter the store. I did not know anything about construction, but I felt
Stimulates me; sexy. really comfortable.
Its like a membership in an exclusive, country-clubish I felt confident and in good hands.
community. Provides the experience that any customer can tackle
American Express any home improvement project.
Its an interactive experience. MasterCard
Part of luxury, sophistication, and exclusivity. Makes me think about precious things in life.
Because of sponsoring activities, I feel fun, excitement, I feel more youthful than using American Express or
and entertainment. Visa.
Apple/iPod Initially the Priceless campaign was emotive, but its
I love the touch and feel of the products. now simply a way of identifying the brand for me.
I enjoy playing with all the products.
I am part of a smarter community. Nike
This brand intrigues me. Makes me think of how to live an active lifestyle.
I really feel Apple products go with my way of life. Makes me feel powerful.
I use the iPod when I am jogging, and I exercise more I want to work out.
because of the iPod. I feel inspired to start working out.
I feel like an athlete.
The Body Shop The store incites me to act, like swing the baseball bat,
Appeals to different senses. or put on the running shoes.
I think of topics like animal testing, purity, and wellness. I enjoy designing my own shoe that perfectly fits my
I want to be with people that share the values that the personality.
brand promotes.
Starbucks
BMW
Smells nice and is visually warm.
I feel young; I feel stylish.
Its just great to drive. Its comfortable and puts me in a better mood.
A BMW is the symbol of my success. Its like being around a Barnes & Nobles crowd.
Crest Target
I feel refreshed. Shopping experience is very pleasant.
Dont really like the smell. Products are displayed to please the eye.
Feels clean, fresh, and healthy. Many stores are putting in Starbucks for an even more
enhanced shopping experience.
Disney
Stimulates my senses. W Hotel
I feel like a child; I feel warm and safe; I want to discover Being part of something fun, happening, and exciting.
things; the brand reminds me to use my imagination. It was an amazing feeling to hang out in the lobby.
I feel part of the magic. Service is disappointing.
Google Washington Mutual
The search is elegant; it creates a mood of playfulness I have positive feelings because of their friendliness.
and curiosity. Its a place I want to go and do not have to go.
I feel happy and proud because I am smart and in-the- I also had a negative one-time experience.
know.
With Google, I change the way I organize and interact Williams-Sonoma
with information. I had a feeling of nostalgia.
Full of memories of home.
HBO Its relaxed and unhurried.
Puts me in a good mood.
Its discussion inducing; I want to discuss the shows
with others.
I enjoy the entertainment.
evoked by the brand on each dimension. As part of the scale highly specific measures for specific experiences, such as
development, we must address methodological challenges. the visual product aesthetics scale (Bloch, Brunel, and
First, in contrast to some other brand scales adopted from Arnold 2003), emotions scales (Izard 1978), or the need-
existing scales in psychology, the development of a brand for-cognition scale (Cacioppo and Petty 1982). These scales
experience scale requires a broader search for acceptable focus on individual reactions to specific stimuli (e.g., well-
items. Second, the items of the scale should focus on the designed products) or tasks (e.g., abstract thinking). They
degree to which a consumer has a sensory, affective, intel- use multifaceted items to measure specific sensations (e.g.,
lectual, behavioral, or social experience with a brand; they I enjoy seeing displays of products that have superior
should not measure the specific sensory, affective, intellec- design; When I see a product that has a really great
tual, behavioral, or social content of the experience (e.g., design, I feel a strong urge to buy it; Bloch, Brunel, and
whether the experience is visually exciting or emotionally Arnold 2003), specific emotions (e.g., affectionate,
warm; what specific imaginary thoughts or behavioral attached, passionate; Thomson, MacInnis, and Park
actions come to mind). Several scales already provide 2005), or specific intellectual processes (e.g., I would pre-
Brand Experience / 57
items that had a loading greater than .4 (88% of items had whether responses to the scale items were truly brand and
loadings greater than .4 on at least one factor). The four- respondent independent and, thus, indicative of a general
factor analysis on the 83 items revealed two factors that brand experience.
were easy to interpret: Factor 3 (sensory experience) and To reduce the number of items further, two independent
Factor 4 (behavioral experience). Factors 1 and 2 included a judges closely examined the 20 items in terms of semantic
mix of items: Factor 1 focused primarily on intellectual similarity. They omitted 8 items (3 sensory, 3 intellectual,
items, and Factor 2 focused on social and affective items. and 2 social items of the social/affective dimension). The
To determine whether the four-factor solution could new scale resulted in 12 items, including 3 items for each of
provide a more distinct structure and to reduce the number the four types of experience dimensions (see Table 2,
of items further, we also conducted an exploratory factor Column 1).
analysis that restricted the number of factors to four (vari- We also prepared a different sample of 30 strong experi-
ance explained = 62%) and then used a stricter loading cri- ential brands (based on expert judgments). We divided them
terion (>.7) to evaluate the Varimax rotated factors. Twenty randomly into six groups of 5 brands each (see Table 3,
items fulfilled the criterion. The results show a distinct pat- Columns 1 and 2). Next, we asked a new sample of 193 stu-
tern: Only sensory items loaded on the first factor (6 items), dents to evaluate the extent to which the 12 items were
only intellectual items loaded on the second factor (6 descriptive of their experience with each brand using a
items), and only behavioral items (2 items had loadings of
greater than .7, and 1 item had a borderline loading of .69) TABLE 2
loaded on the third factor. Finally, both affective and social Study 3: Exploratory Factor Analysis: Revealed
items loaded on the fourth factor (5 items), suggesting that Brand Experience Dimensions
the socially worded items include strong emotional aspects.
In line with Nunnallys (1978) criterion of .7, the Cron- Factor
bachs alphas were satisfactory: They were high for intel- Sensory/
lectual items (.93), sensory items (.93), and social/affective Item Affective Behavioral Intellectual
items (.92), and they were adequate for behavioral items
(.78). This brand makes a
To check the criterion validity of the new, shorter scale, strong impression
on my visual sense
we calculated the mean value of the 20 items for each of the or other senses. .75 .34 .19
21 brands. Ratings on the Apple brand were high and con- I find this brand
sistent (all means ranged from 4.87 to 5.07), indicating that interesting in a
respondents in the five groups used the scale in a similar sensory way. .63 .47 .15
way. Moreover, the data support the criterion validity of the This brand does not
scale. Specifically, all the experiential brands had higher appeal to my
mean values than the nonexperiential brands. The means senses.a .59 .35 .22
This brand induces
of experiential brands ranged from 3.44 to 5.01: 3.44 feelings and
(Reebok), 3.60 (Abercrombie & Fitch), 3.65 (Target), 3.72 sentiments. .82 .13 .27
(W Hotels), 3.86 (Virgin), 4.00 (Coca-Coca), 4.02 (McDon- I do not have strong
alds), 4.09 (Barnes & Noble), 4.10 (JetBlue), 4.22 (Star- emotions for this
bucks), 4.24 (Sony), 4.32 (Tiffany), 4.63 (Nike), 4.67 brand.a .72 .12 .38
(Whole Foods), 4.83 (BMW), and 5.1 (Apple). In contrast, This brand is an
emotional brand. .75 .13 .17
the means of nonexperiential brands were all lower: 2.92
I engage in physical
(Gristedes), 3.08 (Wal-Mart), 3.20 (IBM), 3.27 (Poland actions and
Spring), and 3.33 (Dell). We performed 80 t-tests to exam- behaviors when I
ine statistically whether the mean of each experiential brand use this brand. .18 .80 .22
was significantly different from the mean of each nonexpe- This brand results
riential brand. A total of 75 comparisons (94% of all com- in bodily
parisons) resulted in significant differences (ps < .05). The experiences. .33 .77 .03
This brand is not
nonsignificant comparisons included Reebok and Aber- action oriented.a .13 .72 .27
crombie & Fitch. Specifically, the mean of Reebok was not I engage in a lot of
significantly different from the means of Dell, Poland thinking when I
Spring, and IBM, and the mean of Abercrombie & Fitch encounter this
was not significantly different from the means of Dell and brand. .39 .12 .76
Poland Spring. This brand does not
make me think.a .26 .11 .80
Study 3: Further Item Reduction and Confirmation This brand
stimulates my
of the Dimensions curiosity and
In Study 3, we reduced the number of scale items further problem solving. .15 .24 .75
and conducted both exploratory and confirmatory analyses. aItems are negatively phrased and reverse coded.
To test the stability of the scale, we employed a new sample Notes: Factor analysis uses Varimax rotation. Bold values indicate
of brands and respondents. This enabled us to examine the factor on which each item predominantly loads.
seven-point Likert scale (1 = strongly disagree, and 7 = tual items = .79. In addition, all individual items within
strongly agree). The students were paid $5 each to each dimension averaged item-to-total correlations of .68,
participate. and all exceeded .52, indicating satisfactory levels of inter-
An exploratory factor analysis revealed three factors nal consistency.
with eigenvalues greater than 1. The three factors explained On the basis of these results, we investigated the follow-
67% of the variance. After Varimax rotation, a clean factor ing models further: (1) the three-factor model (sensory/
structure emerged. As Table 2 shows, the 3 sensory and 3 affective, behavioral, and intellectual factors), (2) the four-
affective items loaded on the first factor, the 3 behavioral factor model (sensory, affective, behavioral, and intellectual
items loaded on the second factor, and the 3 intellectual factors), and (3) the nested model (two first-order factors
items loaded on the third factor (see Table 3). behavioral items loading on one factor and intellectual
Because we did not anticipate that both sensory and items loading on anotherplus two other first-order fac-
affective items would load on a single factor, we considered tors, sensory and affective, loading on a second-order
the possibility that these 6 items would load first on two factor).
first-order factors nested within the second-order sensory/ To determine which measurement model fit the data
affective factor. Thus, we conducted another exploratory best, we used structural equation modeling to conduct con-
factor analysis on the 6 sensory and affective items only. firmatory factor analyses. In addition, we analyzed a model
After Varimax rotation, this subsequent analysis revealed that assumed that all items loaded on a single brand experi-
two nested factorsthe sensory factor and the affective fac- ence construct (i.e., one-factor model), a one-factor second-
torwhich explained 74% of the variance. We also order model with four subdimensions, and a one-factor
assessed the internal reliability of the four dimensions of second-order model with three subdimensions. For each
brand experience by calculating a Cronbachs alpha for model, except for the simple one-factor model, we investi-
each dimension using the 12-item scale. All values were gated two cases: a case in which factors were allowed to be
satisfactory: the 3 sensory items = .83, the 3 affective correlated and a case in which the factors were assumed to
items = .81, the 3 behavioral items = .76, and the 3 intellec- be orthogonal. Finally, we grouped the observations by
Brand Experience / 59
brands and centered the means of 12 items at zero to the chi-square statistic in similar models; thus, we consid-
remove any brand effects. ered other fit statistics more valuable in this context. In
The confirmatory factor analyses revealed that the best addition, the ratio between the chi-square statistic and the
model was the four-factor model with correlated factors (for number of degrees of freedom was 5.8, close to 5, indicat-
details, see Figure 1). The fit measures for that model sug- ing an adequate fit (lower values are more desirable; Thom-
gested a reasonable fit: The goodness-of-fit index (GFI) = son, MacInnis, and Park 2005). The fit measures for this
.92, the comparative fit index (CFI) = .91, and the root model were better than those for all other tested models,
mean square error of approximation (RMSEA) = .08, all and the chi-square statistic of this model represented a sig-
indicating acceptable fit, and 2(48) = 278.61, p < .001. nificant improvement over any of the competing models,
Bagozzi and Heatherton (1994) question the usefulness of except for the one-factor second-order model (with four
FIGURE 1
Study 3: Confirmatory Factor Analysis: The Four-Factor Model
.66*
This brand does not appeal to
my senses.a
.81* .70*
This brand induces feelings and
.84*
sentiments.
.67*
This brand is an emotional brand.
.59*
.69*
I engage in physical actions and .77*
behaviors when I use this brand.
.57* .80*
I engage in a lot of thinking when I
.86*
encounter this brand.
.69*
This brand does not make me think.a Intellectual
.62*
This brand stimulates my curiosity and
problem solving.
aReverse coded.
*p < .01.
Notes: All coefficient values are standardized and appear above the associated path. Dotted lines represent correlations.
TABLE 4
Study 3: Confirmatory Factor Analysis Model Fit Comparisons
Null 5079.50 66
One factor 1043.50 54 4036, p< .001
Three factor 527.90 51 515.60, p< .001
One-factor higher order (three subdimensions) 1497.16 48 969.26, p< .001
Nesteda 1209.89 48 287.27, p< .001
Four factor 278.61 48 931.28, p< .001
One-factor higher order (four subdimensions) 278.61 44 0
Brand Experience / 61
from 15 to 56 years; mean age = 23 years) in the shopping evaluation and, in particular, key motivational and affective
streets of two medium-sized cities and asked them to rate brand constructs. Participants (N = 144) completed the 12-
one brand (Nokia) on the 12-item brand experience scale. item brand experience scale and four seven-point general
Second, to assess the discriminant validity of the scale from brand evaluations items (bad/good, unfavorable/
brand attitude, we asked consumers to evaluate the brand on favorable, dislikable/likable, and disagreeable/agree-
three seven-point scales (good/bad, do not like/like very able). In addition, participants completed items related to
much, and not attractive/very attractive). Finally, to the following motivation and affective scales: Zaichkow-
assess the testretest reliability of the scale, two weeks after skys (1985) brand involvement (six items, seven-point
completing the scale for the first time, 72 consumers (mean scale: unimportant to me/important to me, of no concern
age = 22.7 years) volunteered to fill out the brand experi- to me/of concern to me, irrelevant to me/relevant to me,
ence scale for a second time for the same brand. means nothing to me/means a lot to me, useless to me/
The nonstudent sample results were highly similar to useful to me, and insignificant to me/significant to me);
the student sample results in the previous study. Using con- Thomson, MacInnis, and Parks (2005) brand attachment
firmatory factor analysis, the best-fitting model was again a (ten items, seven-point scales: affectionate, loved,
four-factor model with correlated factors (2(48) = 102.85, peaceful, friendly, attached, bonded, connected,
p < .00001; RMSEA = .088). Each item loaded on the pre- passionate, delighted, and captivated); and Oliver,
dicted factor, with standardized coefficients raging from .77 Rust, and Varkis (1997) customer delight (ten items, five-
to .9. point scale: astonished, surprised, contented, happy,
Moreover, the brand experience scale displayed high cheerful, pleased, excited, enthused, stimulated,
discriminant validity from overall brand evaluations. In sev- and elated). To create variation in brand experience
eral exploratory factor analyses, the experience scale and its scores, we manipulated experience strength, asking one-
dimensions were distinct from overall brand evaluations by third of the respondents each to think about a brand that
loading on separate factors. Specifically, a factor analysis provides a strong experience, a moderate experience, or
that did not restrict the number of factors resulted in four a weak, or almost no, experience for them. The manipula-
factors with eigenvalues greater than 1 (variance tion was successful. On the composite experience index,
explained = 74%). After Varimax rotation, Factor 1 had participants felt a stronger experience when asked to report
loadings greater than .68 for the affective and intellectual on a strong brand than when asked to report on a moder-
items, Factor 2 had loadings greater than .79 for the behav- ate or weak experiential brand (M = 5.15, M = 3.84, and
ioral items, Factor 3 had loadings greater than .78 for the M = 2.95, respectively; all ps < .01).
evaluation items, and Factor 4 had loadings greater than .68 To prepare the data for further analyses, we reverse-
for the sensory items. All other loadings were less than .4. coded negatively worded items and computed composite
Another factor analysis (variance explained = 79%) that scores for the four brand experience dimensions (sensory,
restricted the number of factors to 5 (thus examining affective, intellectual, and behavioral), overall brand evalua-
whether the items of the four experience dimensions and tion, brand involvement, the three brand attachment dimen-
the evaluation items would load on different factors) pro- sions (affection, connection, and passion), and customer
vided a similarly distinct rotated factor pattern that could delight. From a factor analysis, we considered customer
easily be interpreted as follows: Factor 1 was behavioral delight a one-dimensional scale.
experience (all loadings > .80), Factor 2 was affective Next, we conducted a factor analysis on the ten compos-
experience (all loadings > .72), Factor 3 was sensory ite brand scales, which revealed a three-factor solution
experience (all loadings > .68), Factor 4 was overall (explaining 75% of the variance). Table 5 shows the factor
brand evaluation (all loadings > .78), and Factor 5 was
intellectual experience (all loadings > .69); all other load- TABLE 5
ings were less than .37. A final factor analysis that restricted Study 5: Exploratory Factor Analysis (Promax
the number of factors to 2 (variance explained = 58%) Rotation), Brand Experience Dimensions, and
resulted in an overall experience factor (all brand experi- Other Brand Scales
ence item loadings > .65) and a separate brand evaluation
factor (all evaluation item loadings > .78); all other loadings Factor
were less than .2.
1 2 3
Finally, we examined the testretest reliability of the
brand experience scale by correlating the ratings at the two Sensory experience .07 .75 .01
measurement points, separated by two weeks. The test Affective experience .21 .48 .32
retest reliability for the overall scale was r = .77, ranging Intellectual experience .26 .82 .13
from r = .69 to r = .73 for the dimensions. Given the time Behavioral experience .23 .58 .54
frame, participants, and environments, the testretest relia- Overall brand evaluation .19 .14 .81
Brand involvement .05 .33 .68
bility seems adequate. Affection .55 .00 .45
Connection .44 .02 .58
Study 5: Discriminant Validity of the Brand Passion .73 .23 .05
Experience Scale Customer delight .88 .02 .00
We conducted Study 5 to provide additional evidence for Notes: Standardized regression coefficients. Bold values indicate
the discriminant validity of the scale in relation to brand the factor on which each item predominantly loads.
Brand Experience / 63
cations. A useful input in this inference is likely to be brand class and charming (for sophistication); and outdoorsy
experience. A trait judgment about a brands sincerity, and tough (for ruggedness) (Aaker 1997, p. 352). We
excitement, competence, sophistication, or ruggedness can measured items on a seven-point Likert scale (1 = not at all
be facilitated when the consumer attends to specific sen- descriptive, and 7 = extremely descriptive) and provided
sory, affective, intellectual, or behavioral experiences. For the instructions in line with the work of Aaker (1997).
example, to conclude that Hallmark is sincere (Aaker We measured consumer satisfaction using five items
1996), a consumer may attend to his or her feelings of hap- modeled after Oliver (1980): I am satisfied with the brand
piness, romance, or nostalgia or thoughts of holiday activi- and its performance, and If I could do it again, I would
ties. Similarly, to conclude that a clothing brand, such as buy a brand different from that brand (negative item,
Levis, is rugged (Aaker 1996), a consumer may attend to reverse coded); My choice to get this brand has been a
his or her sensory experiences based on the brands colors, wise one, and I feel bad about my decision to get this
thoughts about the Wild West origin of the brand, or bod- brand (negative item, reverse coded); and I am not happy
ily experiences based on the fit and texture of the jeans. with what I did with this brand (negative item, reverse
Thus, experiences are used as information (Pham 2004). coded). Each seven-point Likert scale was anchored by
This information, in conjunction with other information, strongly disagree (1) and strongly agree (7).
may be combined into a brand personality judgment. Finally, to measure consumer loyalty, we adopted five
As a result, we expect that brand experience is an standard loyalty items from the work of You and Donthu
antecedent of brand personality. The higher the overall (2001): In the future, I will be loyal to this brand; I will
score on the brand experience scale, the more likely the buy this brand again; This brand will be my first choice in
consumer will endow the brand with personality associa- the future; I will not buy other brands if this brand is
tions. The reverse processbrand personality preceding available at the store; and I will recommend this brand to
brand experienceis conceptually less plausible; it is not others. Again, we measured the items on a seven-point
clear how sensory, affective, intellectual, or behavioral Likert scale ranging from strongly disagree (1) to
experiences could easily result from a summary judgment strongly agree (7).
such as brand personality. Participants rated a set of 12 brands for six categories:
Apple and Dell (computers), Fiji and Poland Spring (water),
H3: Brand experience affects brand personality positively. J. Crew and Liz Clairborne (clothing), Puma and New Bal-
Brand personality provides differentiation, increases ance (sneakers), Volkswagen and Saturn (cars), and the New
preference, and enhances trust and loyalty (Biel 1993; York Times and USA Today (newspapers). Each participant
Fournier 1998). Moreover, selecting a brand with a certain evaluated two categories and both brands within that cate-
personality enables consumers to express themselves gory. We counterbalanced the order of presentation of the
(Aaker 1999). Thus, brand personality offers value to con- category and the order of brands within each category.
sumers similar to experiences. Therefore, the more a brand
is associated with human characteristics, the more satisfied
Results and Discussion
and loyal a consumer will be. Before estimating the structural equation model based on
our conceptual model, we examined the discriminant
H4: Brand personality affects consumer satisfaction positively.
validity of the brand experience scale from the brand per-
H5: Brand personality affects consumer loyalty positively.
sonality scale. Given the large sample size, we were able to
Finally, it has been shown that consumer satisfaction conduct an exploratory factor on the entire set of original
affects loyalty. When a consumer feels good about the rela- itemsthe 12 items of the brand experience scale and the
tionship and appreciates the product or brand, a high level 15 items of the short version of the brand personality scale
of commitment and loyalty results (Anderson and Sullivan that address the five dimensions of brand personality (sin-
1993; Mittal and Kamakura 2001; Oliver 1997). Thus: cerity, excitement, competence, sophistication, and rugged-
ness). The exploratory factor analysis revealed five factors
H6: Consumer satisfaction affects consumer loyalty positively.
with eigenvalues greater than 1, but the scree plot exhibited
a significant dip between the fourth and the fifth factor. The
Procedure first four factors explained 62% of the variance. After Vari-
A total of 209 students participated in Study 6 for a com- max rotation, a clean factor structure emerged (see Table 6).
pensation of $5. Each participant rated the extent to which In general, brand experience and brand personality exhib-
the items described his or her experiences with the brands ited high levels of discriminant validity: The respective
listed, the personality of the brands listed, and feelings of items loaded on separate factors. Factor 3 was the only fac-
satisfaction and loyalty toward the brands. tor that included both personality and brand experience
The brand experience scale included the 12-item scale items, namely, behavioral experience and ruggedness items.
we used in Studies 3, 4, and 5. To measure brand personal- We also conducted a factor analysis on the composite
ity, we included a version of the scale that consisted of the scores of each brand experience and brand personality
15 brand personality items that represented the five brand dimension, which further confirmed the discriminant
personality dimensions: down-to-earth, honest, whole- validity of the scales. Two factors had eigenvalues greater
some, and cheerful (for sincerity); daring, spirited, than 1. After Varimax rotation, the experience dimensions
imaginative, and up-to-date (for excitement); reliable, and the personality dimensions loaded separately on the two
intelligent, and successful (for competence); upper- factors; however, on the three-factor solution, the behav-
Brand Experience / 65
FIGURE 2
Study 6: Discriminant and Predictive Validity of the Brand Experience Scale
Brand
Personality
Sensory .86*
.69* .67* .13 * *
Affective .81*
Brand .15** .59*
Experience Satisfaction Loyalty
.81*
Intellectual
.60*
Behavioral .24*
*p < .01.
**p < .05.
Notes: All coefficient values are standardized and appear near the associated path.
worded and negatively worded versions of the scale and rants, spas, airline flights), extend over time. Research has
investigate how positive and negative experiences affect shown that the pattern of extended experiences (e.g.,
consumer behavior. There seems to be two ways to create increasing or decreasing in intensity) can affect how experi-
valenced versions of the scale. One way is to reword the ences are evaluated (Ariely 1998; Ariely and Zauberman
scale items slightly by including the words positive or 2003). Such research has used the term experience gener-
negativefor example, This brand makes a strong ically to refer to responses within a certain time frame.
positive/negative impression on my visual sense or other From the studies reported here, it would be worthwhile to
senses, This brand induces positive/negative feelings and explore which patterns occur for which experience dimen-
sentiments, This brand results in positive/negative bodily sions. Moreover, we assessed experiences retrospectively;
experiences, or I engage in a lot of positive/negative we did not directly assess a consumers dynamic experience
thinking when I encounter this brand. The second way is to with a brand. Further research should investigate whether
add a phrase and a bipolar response scale after each scale retrospective experiences are different from dynamic expe-
itemfor example, and this results in an experience riences, in terms of both structure (i.e., the type of dimen-
that is very positive (1) or very negative(7). Further sions that are most commonly involved) and content (i.e.,
research should determine which method is more reliable the specific sensations, feelings, cognitions, and behaviors
and valid. that are involved).
Further research should also examine whether the scale Finally, further research should focus on the antecedents
can predict specific behavioral outcomes. On the basis of and long-term consequences of brand experiences. For
our theorizing, we would expect the sensory dimension to example, regarding antecedents, how exactly are brand
predict future design and aesthetics perceptions and usages, experience dimensions evoked by brand-related stimuli? In
the affective experience to predict emotional judgments, the addition, although we have shown the impact of brand
intellectual dimension to predict creative usages of the experience, both directly and indirectly, on short-term con-
brand, and the behavioral experience to predict specific sequences, such as satisfaction and loyalty, the question
actions and physiological reactions when interacting with a arises whether brand experiences affect customer lifetime
brand. value (Rust, Zeithaml, and Lemon 2000; Vogel, Evan-
In addition to research on the scale, we encourage schitzky, and Ramaseshan 2008). That is, can brand experi-
research on the experience concept. For example, many ences build customer equity, and how should marketers
experiences, especially in the services industry (e.g., restau- manage brands to create experiences that build such equity?
Brand Experience / 67
Holt, Douglas B. (1995), How Consumers Consume: A Typology (1993), Cognitive, Affective, and Attribute Bases of the
of Consumption Practices, Journal of Consumer Research, 22 Satisfaction Response, Journal of Consumer Research, 20
(June), 116. (December), 41830.
Huffman, Cynthia and Michael J. Houston (1993), Goal-Oriented (1997), Satisfaction: A Behavioral Perspective on the
Experiences and the Development of Knowledge, Journal of Consumer. Boston: McGraw-Hill.
Consumer Research, 20 (September), 190207. , Roland T. Rust, and Sajeev Varki (1997), Customer
Hui, Michael K. and John E.G. Bateson (1991), Perceived Con- Delight: Foundations, Findings, and Managerial Insight, Jour-
trol and the Effects of Crowding and Consumer Choice on the nal of Retailing, 73 (3), 31136.
Service Experience, Journal of Consumer Research, 18 (Sep- Park, C. Whan and Deborah J. MacInnis (2006), Whats In and
tember), 17484. Whats Out: Questions over the Boundaries of the Attitude
Izard, Carroll E. (1978), Human Emotions, 2d ed. New York: Construct, Journal of Consumer Research, 33 (1), 1618.
Plenum.
Parsons, Andrew and Denise Conroy (2006), Sensory Stimuli and
Johar, Gita, Jaideep Sengupta, and Jennifer Aaker (2005), Two
E-Tailers, Journal of Consumer Behaviour, 5 (1), 6981.
Roads to Updating Brand Personality Impressions: Trait Versus
Evaluative Inferencing, Journal of Marketing Research, 42 Petty, Richard E. and John T. Cacioppo (1986), The Elaboration
(November), 45869. Likelihood Model of Persuasion, in Advances in Experimental
Jones, Monique A. (1999), Entertaining Shopping Experiences: and Social Psychology, Vol. 19, Leonard Berkowitz, ed. New
An Exploratory Investigation, Journal of Retailing and Con- York: Academic Press, 123205.
sumer Services, 6 (3), 12939. Pham, Michel (2004), The Logic of Feelings, Journal of Con-
Joy, Annamma and John F. Sherry Jr. (2003), Speaking of Art as sumer Psychology, 14 (4), 36069.
Embodied Imagination: A Multisensory Approach to Under- Pine, Joseph B., II, and James H. Gilmore (1999), The Experience
standing Aesthetic Experience, Journal of Consumer Economy: Work Is Theatre and Every Business a Stage. Cam-
Research, 20 (September), 25982. bridge, MA: Harvard Business School Press.
Keller, Kevin L. (1987), Memory Factors in Advertising: The Pinker, Steven (1997), How the Mind Works. New York: Norton.
Effects of Advertising Retrieval Cues on Brand Evaluations, Plutchik, Robert and Henry Kellerman, eds. (1989), Emotion:
Journal of Consumer Research, 14 (December), 31633. Theory, Research, and Experience. San Diego: Academic
(1993), Conceptualizing, Measuring, and Managing Press.
Customer-Based Brand Equity, Journal of Marketing, 57 (Jan- Reicheld, Fredrick (1996), The Loyalty Effect: The Hidden Force
uary), 122. Behind Growth, Profits, and Lasting Value. Boston: Harvard
Kempf, Deanna S. and Robert E. Smith (1998), Consumer Pro- Business School Press.
cessing of Product Trial and the Influence of Prior Advertising: Richins, Marsha L. (1997), Measuring Emotions in the Con-
A Structural Modeling Approach, Journal of Marketing sumption Experience, Journal of Consumer Research, 24
Research, 35 (August), 32538. (September), 12746.
Kerin, Roger A., Ambuj Jain, and Daniel J. Howard (1992), Store Rust, Roland T., Valarie A. Zeithaml, and Katherine N. Lemon
Shopping Experience and Consumer PriceQualityValue Per-
(2000), Driving Customer Equity: How Customer Lifetime
ceptions, Journal of Retailing, 68 (4), 37697.
Value Is Reshaping Corporate Strategy. New York: The Free
Lakoff, George and Mark Johnson (1999), Philosophy in the
Press.
Flesh: The Embodied Mind and Its Challenge to Western
Thought. New York: Basic Books. Schmitt, Bernd H. (1999), Experiential Marketing: How to Get
Mandel, Naomi and Eric J. Johnson (2002), When Web Pages Customers to Sense, Feel, Think, Act, Relate to Your Company
Influence Choice: Effects of Visual Primes on Experts and and Brands. New York: The Free Press.
Novices, Journal of Consumer Research, 29 (September), (2003), Customer Experience Management. New York:
23545. John Wiley & Sons.
McAlexander, James H., John W. Schouten, and Harold F. Koenig and Alex Simonson (1997), Marketing Aesthetics: The
(2002), Building Brand Community, Journal of Marketing, Strategic Management of Brands. New York: The Free Press.
66 (January), 3854. Shaw, Colin and John Ivens (2002), Building Great Customer
McAllister, Leigh and Edgar A. Pessemier (1982), Variety Seek- Experiences. New York: Palgrave/Macmillan.
ing Behavior: An Interdisciplinary Review, Journal of Con- Smith, Robert E. and Xiaojing Yang (2004), Toward a General
sumer Research, 9 (December), 31122. Theory of Creativity in Advertising: Examining the Role of
Meyers-Levy, Joan and Laura A. Peracchio (1995), How the Use Divergence, Marketing Theory, 4 (12), 3158.
of Color in Advertising Affects Attitudes: The Influence of Pro- Smith, Shaun and Joe Wheeler (2002), Managing the Customer
cessing Motivation and Cognitive Demands, Journal of Con- Experience: Turning Customers into Advocates. Upper Saddle
sumer Research, 22 (September), 12138. River, NJ: Financial Times/Prentice Hall.
Mittal, Vikas and Wagner A. Kamakura (2001), Satisfaction, Solomon, Michael R. (2004), Consumer Behavior: Buying, Hav-
Repurchase Intent, and Repurchase Behavior: Investigating the ing, and Being, 6th ed. Upper Saddler River, NJ: Pearson.
Moderating Effect of Customer Characteristics, Journal of Thomson, Matthew, Deborah J. MacInnis, and C. Whan Park
Marketing Research, 38 (February), 13142. (2005), The Ties That Bind: Measuring the Strength of Con-
Muniz, Albert M., Jr., and Thomas C. OGuinn (2001), Brand sumers Emotional Attachments to Brands, Journal of Con-
Community, Journal of Consumer Research, 27 (March),
sumer Psychology, 15 (1), 7791.
41232.
Murphy, Sheila T. and R.B. Zajonc (1993), Affect, Cognition and Veryzer, Robert W. and J. Wesley Hutchinson (1998), The Influ-
Awareness: Affective Priming with Optimal and Suboptimal ence of Unity and Prototypicality on Aesthetic Responses to
Stimulus Exposures, Journal of Personality and Social Psy- New Product Designs, Journal of Consumer Research, 24
chology, 64 (5), 72339. (March), 37494.
Nunnally, Jum C. (1978), Psychometric Theory. New York: Vogel, Verena, Heiner Evanschitzky, and B. Ramaseshan (2008),
McGraw-Hill. Customer Equity Drivers and Future Sales, Journal of Mar-
Ofir, Chezy and Itamar Simonson (2007), The Effect of Stating keting, 72 (November), 98108.
Expectations on Consumer Satisfaction and Shopping Experi- You, Xueming and Naveen Donthu (2001), Developing and Vali-
ence, Journal of Marketing Research, 44 (February), 16474. dating a Multidimensional Consumer-Based Brand Equity
Oliver, Richard L. (1980), A Cognitive Model of the Antecedents Scale, Journal of Business Research, 52 (April), 114.
and Consequences of Satisfaction Decisions, Journal of Mar- Zaichkowsky, Judith L. (1985) Measuring the Involvement Con-
keting Research, 17 (November), 46069. struct, Journal of Consumer Research, 12 (3), 34152.